Canadian Science Advisory Secretariat (CSAS)
Research Document 2015/039
Pacific Region
Trends in the abundance and distribution of sea otters (Enhydra lutris) in British
Columbia updated with 2013 survey results
L. M. Nichol1, J.C. Watson2, R. Abernethy1, E. Rechsteiner3, J. Towers1
1
Fisheries & Oceans Canada,
Pacific Biological Station,
3190 Hammond Bay Rd.,
Nanaimo, BC, V9T 6N7
2
Vancouver Island University,
900 Fifth Street,
Nanaimo, BC, V9R 5S5
3
Hakai Beach Institute,
Box 309
Heriot Bay, BC, V0P 1H0
July 2015
Foreword
This series documents the scientific basis for the evaluation of aquatic resources and
ecosystems in Canada. As such, it addresses the issues of the day in the time frames required
and the documents it contains are not intended as definitive statements on the subjects
addressed but rather as progress reports on ongoing investigations.
Research documents are produced in the official language in which they are provided to the
Secretariat.
Published by:
Fisheries and Oceans Canada
Canadian Science Advisory Secretariat
200 Kent Street
Ottawa ON K1A 0E6
http://www.dfo-mpo.gc.ca/csas-sccs/
[email protected]
© Her Majesty the Queen in Right of Canada, 2015
ISSN 1919-5044
Correct citation for this publication:
Nichol, L.M., Watson J.C., Abernethy, R, Rechsteiner, E., Towers, J. 2015. Trends in the
abundance and distribution of sea otters (Enhydra lutris) in British Columbia updated with
2013 survey results. DFO Can. Sci. Advis. Sec. Res. Doc. 2015/039. vii + 31 p.
Table of Contents
ABSTRACT ................................................................................................................................ vi
RÉSUMÉ .................................................................................................................................. vii
INTRODUCTION ........................................................................................................................1
METHODS ..................................................................................................................................3
Survey Area ........................................................................................................................... 3
Survey Method ....................................................................................................................... 4
Analytical Methods ................................................................................................................. 4
Maps .................................................................................................................................. 6
RESULTS ...................................................................................................................................6
Regional Population Estimates and Growth Rates ................................................................. 6
Vancouver Island Region ................................................................................................... 6
Central Mainland Coast Region .......................................................................................... 6
Sub-regional Annual Rates of Increase and range expansion ................................................ 7
Vancouver Island Region ................................................................................................... 7
Central Mainland Coast ...................................................................................................... 7
DISCUSSION..............................................................................................................................8
ACKOWLEDGEMENTS ............................................................................................................11
REFERENCES .........................................................................................................................11
TABLES ....................................................................................................................................13
FIGURES ..................................................................................................................................17
iii
List of Tables
Table 1. Kilometres of survey effort and survey conditions in 2008 and 2013. ..........................13
Table 2. Sea otter counts by region in years when at least one of the regions was surveyed
entirely*. Platforms : fixed-wing aircraft- 1, helicopter- 2, small boat- 3. Coast Guard
ship - 4. BC Total is only given when both regions were surveyed. ...................................14
Table 3. Model fitting results for sub-regions of Vancouver Island illustrated in Figure 13. ........15
Table 4. Model fitting results for sub-regions of the central mainland coast, illustrated in
Figure 15. ..........................................................................................................................16
List of Figures
Figure 1. Survey region boundaries and areas within that were surveyed in 2008 and 2013.
Checleset Bay site of the original re-introductions 1969-1972. Goose Group where
groups of sea otters were first sighted on the central mainland coast in 1989. ..................17
Figure 2. Survey segments and new areas surveyed in 2013. Survey segments: 1. Clayoquot
Sound, 2. Hesquiat Peninsula, 3. Nootka Island, 4. Nuchatlitz Inlet, 5. Catala Island, 7.
Kyuquot Sound, 6. Esperanza Inlet, 8. Mission Group, 9. Checleset Bay, 10. Brooks
Bay, 11. Quatsino Sound, 12. Kains Island to Cape Scott, 13. Scott Islands , 14. Cape
Scott to Hope Island, 15. Queen Charlotte Strait east, 16. Calvert Island, 17. Kildidt
Sound, 18. Simond Group to Tribal Group, 19. Goose Group, 20. McMullin Group to
Cape Mark, 21. Seaforth to Ivory Island and Lady Douglas Island, 22. Aristazabal
Island. New areas surveyed to assess range expansion: a. Inlets connected to
Esperanza Inlet, b. Holberg Inlet, c. Smith Sound, d. Price Island, e. extension of
Aristazabal Island area. .....................................................................................................18
Figure 3. Sea otter range in British Columbia in 2013 and place names mentioned in the text. .19
Figure 4. Distribution of sea otter sightings and survey effort on northern Vancouver Island in
2013. .................................................................................................................................20
Figure 5. Distribution of sea otter sightings and survey effort on central west coast Vancouver
Island in 2013. ...................................................................................................................21
Figure 6. Distribution of sea otter sightings and survey effort on the central mainland coast in
2013. .................................................................................................................................22
Figure 7. Distribution of sea otter sightings and survey effort on northern Vancouver Island in
2008. .................................................................................................................................23
Figure 8. Distribution of sea otter sightings and survey effort on the central west coast of
Vancouver Island in 2008. .................................................................................................24
Figure 9. Distribution of sea otter sightings and survey effort on the central mainland coast in
2008. .................................................................................................................................25
Figure 10. Number of sea otters by group sizes that comprised the 6,754 sea otters counted
during surveys in 2013. .....................................................................................................26
Figure 11. A. Estimated population growth curve in the Vancouver Island region 1977 to
2013 on a log scale. Solid line represents the modelled expected values. Dotted line
represents approximate 95% confidence intervals. Diamonds represent survey counts
from years when all occupied survey segments were surveyed. B. Growth curve and
95% confidence intervals and survey count on an ordinal scale. .......................................27
iv
Figure 12. A. Estimated population growth curve on the central mainland coast region 1990
to 2013 on a log scale. Solid line represents the modelled expected values. Dotted line
represents approximate 95% confidence intervals. Diamonds represent survey counts
from years when all occupied survey segments were surveyed. B. Growth curve and
95% confidence intervals and survey count on an ordinal scale. .......................................28
Figure 13. Estimated growth trends in the Vancouver Island region. Inset graphs show
growth curves by subregion, either logistic or exponential (on a log scale) fitted to sea
otter survey data. Inset pie chart gives the proportions of the 2013 count ( 5,612 sea
otters) by sub-region. ........................................................................................................29
Figure 14. Pattern of colonization and growth observed in the Catala Island segment
following initial range expansion by a raft of males in 1992 (red triangle). Diamonds
represent counts in subsequent years which were predominantly females and pups. .......30
Figure 15. Estimated growth trends in the central mainland coast region. Inset graphs show
growth curves by subregions either logistic or exponential (on log scale) fitted to sea
otter survey data. Inset pie chart gives the proportions of the 2013 count (1,142 sea
otters) by sub-region. ........................................................................................................31
v
ABSTRACT
In 2013 the BC sea otter population had a minimum of 6,754 sea otters: 5,612 sea otters were
counted in the Vancouver Island region and 1,142 sea otters in the central mainland coast
region. The population has continued to grow and expand its range since the last survey,
demonstrating population growth patterns that are typical among recovering sea otter
populations. In the Vancouver Island region, which encompasses the west and north coast of
Vancouver Island as well as Queen Charlotte Strait, annual growth was estimated regionally to
have been 6.77% year-1 (SE = 1.20) during 2004 to 2008 and 7.12% year-1 (SE = 1.05) from
2009 to 2013. These rates are similar but slightly lower than previous estimates for the period
1995 to 2008. They reflect density dependent growth along some parts of Vancouver Island
where the sea otter population is approaching carrying capacity. This trend was detected in this
assessment by fitting growth models to sub-regional data sets and estimating growth rates over
shorter time intervals than has been done previously. On the central BC mainland coast, growth
averaged 4.22% year-1 (SE = 1.62) during 2004 to 2008 reflecting density dependent growth in
the longest occupied sub-region. However the contribution of exponential growth in sub-regions
where range expansion was first noted in 2007 and 2008 contributed to a much higher regional
rate of 12.56% year-1 (SE = 1.65) for the most recent five-years 2009 to 2013.
vi
Tendances relatives à l'abondance et à la répartition de la loutre de mer (Enhydra
lutris) en Colombie-Britannique mises à jour d'après les résultats du relevé de
2013
RÉSUMÉ
En 2013, la population de loutres de mer en Colombie-Britannique comptait au moins 6 754
individus : 5 612 loutres de mer ont été dénombrées dans la région de l'île de Vancouver et
1 142 sur la côte centrale de la province. La population a continué de croître et étendu son aire
de répartition depuis le dernier relevé, suivant des profils de croissance caractéristiques des
populations de loutres de mer en voie de rétablissement. Dans la région de l'île de Vancouver,
qui englobe les côtes ouest et nord de l'île de Vancouver ainsi que le détroit de la ReineCharlotte, la croissance annuelle estimée à l'échelle régionale était de 6,77 % par année-1
(ET = 1,20) de 2004 à 2008 et de 7,12 % par année-1 (ET = 1,05) de 2009 à 2013. Ces taux
sont similaires, quoique légèrement inférieurs, aux estimations précédentes pour la période
s'étendant de 1995 à 2008. Ils reflètent la dépendance à la densité à certains endroits sur le
littoral de l'île de Vancouver là où la population se rapproche de sa capacité de charge. Cette
tendance a été observée dans cette évaluation en ajustant les modèles de croissance aux
ensembles de données sous-régionales et en estimant les taux de croissance pour des
intervalles plus courts que ce qui avait été fait auparavant. Sur la côte centrale de la ColombieBritannique, la croissance moyenne était de 4,22 % par année-1 (ET = 1,62) de 2004 à 2008,
situation qui témoigne de la dépendance à la densité dans la plus longue sous-région
fréquentée par la loutre de mer. Toutefois, la croissance exponentielle dans les sous-régions où
l'élargissement de l'aire de répartition a été noté pour la première fois en 2007 et 2008 a
contribué à un taux de croissance régional beaucoup plus élevé de 12,56 % par année-1
(ET = 1,65) au cours des cinq dernières années (2009 à 2013).
vii
INTRODUCTION
Knowledge of population abundance and growth trends is essential for conservation of sea
otters in British Columbia. Canada’s sea otter population includes descendants of animals
reintroduced to Checleset Bay on the west coast of Vancouver Island from Alaska during three
translocation efforts in 1969, 1970 and 1972 (Bigg and MacAskie 1978). The original population
is thought to have been extirpated by 1930 as a result of commercial hunting for pelts that had
begun with the maritime fur trade along the west coast of North America beginning in the late
1700’s (Nichol 2015). Following successful reintroduction of 89 animals from Alaska, the BC
sea otter population was first surveyed to estimate abundance in 1977 and subsequently at 1 - 3
year intervals thereafter, although not all the range was surveyed. In 1989 a group of sea otters
were reported from the Goose Island Group on the central mainland coast signifying the
establishment of sea otters in the area (BC Parks 1995 1).
From 1977 to 1988, surveys were made mostly from fixed-wing aircraft (Bigg and MacAskie
1978; Morris et al. 1981; MacAskie 1987). Since 1988 a standardized survey approach has
been used to survey by small boat. The method is a direct count of sea otters in their known
range whereby the range is surveyed by the same routes, using consistent methods with the
same observers or observers with comparable experience from survey to survey. The precision
of replicate counts obtained by this survey method in Good to Excellent survey conditions has
previously been estimated to be (CV = 7-12%) indicating that a high degree of precision is
achievable (Watson et al. 1997; Nichol et al. 2005). The resulting counts provide an index of
abundance and data with which to calculate trends in population growth. In 2001 the method
was modified to include helicopters as an alternative platform to facilitate coverage of large and
remote coastal areas (Nichol et al. 2005). Direct count approaches are used to obtain
population index counts and estimate average growth rates in the known range in Washington
State and California as well, but they use fixed wing aircraft and shore counts (Jameson and
Jeffries 2014; USGS 2014).
Sea otter populations can be successfully surveyed using these methods because of specific
characteristics of sea otter behaviour and biology that make them predictable in distribution.
Sea otters live in nearshore coastal habitats and are most commonly found within the 40-metre
depth contour. The sea otter feeds on benthic invertebrates which they collect from the sea floor
by diving. Sea otters require frequent access to foraging habitat in subtidal and intertidal habitat
and most foraging dives occur in depths of 30 metres or less (Reidman and Estes 1990; Bodkin
et al. 2004; Lafferty and Tinker 2014). The extent of sea otter habitat is thus set by the limits of
their preferred foraging depth with the result that most sea otters are found within 1 to 2 km of
shore unless shallow areas extend further offshore (Reidman and Estes 1990; Bodkin et al.
2004).
Within this relatively narrow nearshore band, sea otters are non-migratory and occupy small
overlapping home ranges throughout their lives. Territorial male home ranges can be 4 to 11
km2, and female home ranges vary from a few to 24 km2 (Garshelis and Garshelis, 1984; Ralls
et al. 1988; Jameson, 1989). The sea otter is gregarious and spends a considerable amount of
its time resting in floating aggregations called rafts, that may number over 200 animals. Sea
otters segregate by sex and form male rafts and female and pup rafts. Rafts form habitually in
the same location and are often associated with particular reefs and kelp beds. The aggregating
behaviour of sea otters results in a clumped distribution that is probably also associated with
1
BC Parks. 1995. Sea Otter Management Plan, for Hakai Recreation Area. [Unpublished
reported prepared by J.C. Watson for] British Columbia Ministry of Water Land and Air
Protection, Parks and Protected Areas Branch.
1
habitat characteristics. In British Columbia, sea otters generally occupy coastal areas exposed
to the open Pacific as opposed to protected water ways. Along the exposed coast, sea otters
are found particularly along shorelines characterized by jagged coastlines with clusters of small
islets and reefs and shallow variable depths (Gregr et al. 2008).
The distribution of rafts of sea otters can be an indicator of range expansion events in growing
populations. The periphery, or frontal edge of the occupied range, tends to be occupied by male
rafts. Range expansion in a growing population is triggered by population density in the
occupied range. The appearance of a male raft in a previously unoccupied area signifies the
first phase of range expansion. In subsequent years, rafts of females form in the new area
(Garshelis et al. 1984; Jameson 1989; Lafferty and Tinker 2014).
At the time of sea otter re-introductions to British Columbia, sea otters had also been
reintroduced to Washington State and to Southeast Alaska with success (Jameson et al. 1982).
These reintroduced populations, including the one in BC, grew rapidly in the early years at rates
of 17-25% per year, the maximum rate possible for the species in the absence of density
dependent factors (Estes 1990). The reason for these high rates during the early phase was
likely a result of abundant invertebrate prey which had increased in the absence of sea otters.
Sea otter populations are regulated by food through density dependent processes that result in
elevated rates of mortality, particularly juvenile mortality, as food becomes limiting in an area
(Estes 1990; Thometz et al. 2014). The minimum population estimate for the Washington State
population in 2013 was 1,272 animals with growth since 1989 of 7.6% year-1 whereas growth
from 1978 to 1987 was estimated to be 17.2% year-1 (Estes 1990; Jameson and Jeffries 2014).
In Southeast Alaska, an aerial strip transect method is used (Bodkin and Udevitz 1999). The
most recent population estimate from surveys (2005 to 2012) is 25,712 otters (based on a total
of five sub-region estimates with CVs ranging from 0.18 to 0.38) with an annual growth rate of
12 to 14% year-1 (USFW 2014). The growth rate in Southeast Alaska estimated from 1975 to
1987 is 17.6% year-1 (Estes 1990).
Following reintroduction of sea otters into Checleset Bay (in 1969-72), the population grew but
until 1987 was found in only two locations along the west coast of Vancouver Island, Checleset
Bay and off Nootka Island (Bigg and MacAskie 1978). By 1995, the population had increased to
1,522 otters and was distributed along Vancouver Island from Estevan Point to the entrance of
Quatsino Sound (Watson et al. 1997). By 2004 the population was distributed from Clayoquot
Sound to the northwestern edge of Queen Charlotte Strait. On the central mainland coast the
population had expanded northeast ward from the Goose Group to the edge of Milbanke Sound
by 2001 and to Aristazabal Island by 2008 (Nichol et al. 2009). By 2008 when a range-wide
survey was completed, the sea otter population in BC included a minimum of 4,712 sea otters,
with 4,110 occupying the Vancouver Island region and 602 the central mainland coast region
(Nichol et al. 2009). The growth rate on Vancouver Island was estimated to be 19.0% year-1
from 1977 to 1995 but slowed thereafter to 8.4% year-1 to 2008. Population growth on the
central mainland coast averaged 11.4% year-1 from 1990 to 2008.
In 2013, a range-wide boat survey was completed which encompassed the occupied range on
Vancouver Island and the central mainland coast. As well, most of the central mainland coast
range was also surveyed in 2011 by boat. In this report, an new index count for 2013 is
presented along with population growth rate estimates from within and between the two regions
of Vancouver Island and the central mainland coast.
The Terms of Reference specifically for this document are;
•
Describe changes in the range of the sea otter in 2013 compared to previous
assessments and compared to the historic range and since re-introduction in 1969-72.
•
Provide a minimum abundance estimate for the population in 2013,
2
•
Describe changes in abundance that have occurred since 2008, since 1969-72 and
compared to historic abundance.
•
Estimate the population growth trend to 2013 in total and by area and compare to 2008
and since reintroduction in 1969-72.
To address the specific request above regarding growth trends by area, growth rates were
estimated for sub-regions that comprise the Vancouver Island range and the central mainland
coast.
METHODS
SURVEY AREA
For the assessment, two regions are defined that encompass all coastal areas surveyed. The
Vancouver Island region includes the entire occupied range on Vancouver Island as well as
Queen Charlotte Strait and the adjoining portion of the mainland. The Vancouver Island region
includes Checleset Bay where sea otters were reintroduced in 1969, 1970 and 1972 and thus
where sea otters have been established for the longest period of time (Figure 1). The second
region, called the central mainland coast, includes the coastal waters from Calvert Island north
to Aristazabal Island. The central mainland coast region includes the Goose Group, where a raft
of sea otters was first reported in 1989.
To facilitate systematic survey coverage, the sea otter range within these regions was divided
into segments. Segments boundaries are natural breaks such as points of land, or the transition
from a sound to an inlet. Segments can typically be surveyed in a day by boat, although some
segments delineated in the past now take more than one day to complete because of population
growth, as it takes longer to count more otters. Within segments, survey coverage is defined by
established survey routes. As of 2013 16 segments comprise the Vancouver Island region and a
further six segments comprise the central mainland coast region (Figure 2).
In addition to survey segments, areas adjacent to survey segments or areas from which a report
had been received of sea otters were also surveyed to monitor range expansion. Over time new
survey segments were added as the sea otter range expanded. The marine mammal program
at the Pacific Biological Station solicits reports of sea otter sightings from other researchers,
fisheries patrol vessel personnel, fisheries enforcement officers, coastal residents and
fishermen to help identify range expansion events and thereby define the extent of survey
coverage needed during the range-wide surveys completed at five year intervals. Since 2001,
surveys have also been made periodically in inlets adjacent to several occupied areas such as
Clayoquot Sound, Checleset Bay, Esperanza Inlet and Quatsino Sound on Vancouver Island, to
ascertain whether there has been unreported movement of otters into these areas.
Following the survey of new areas, only those in which a raft of sea otters was observed during
a survey were considered occupied. This criterion is used, for consistency, to identify range
expansion events. Although single sea otters are encountered and reported outside the
occupied range occasionally throughout BC (Ford 2014), the presence of rafts during spring or
summer – which is when sea otter surveys are undertaken – is used to define range of
occupation. During winter months it is not uncommon for a raft of sea otters to make irregular
appearances in new areas, but consistent occupation of the new area may take several years
and coincides with observed occupation during spring and summer. Survey coverage in 2013,
including new areas surveyed in 2013, is shown in Figure 1.
3
SURVEY METHOD
Survey routes within segments followed the coastline and included coverage of waters around
all islets and reefs; habitats typically occupied by sea otters. Survey routes were recorded by
interfacing a GPS device with navigation software (Nobeltec Visual Navigation Suite, Nobeltec
Corporation, or Navionics) on a laptop or iPad. Small boat surveys were conducted by two or
three observers and one boat driver. The small vessels used were 5.5-metre welded aluminum
boats or 6.5-metre rigid hull inflatable boats. Observers searched for and counted sea otters on
either side as well as forward of the boat as the vessel proceeded along the course. The
number of otters sighted was recorded and their location was approximated as the position of
the boat at the time of the sighting. Locations of rafts were further refined by examining the
marine features on the chart and the boat’s position. The boat traveled at speeds of less than 10
knots (18.5 km/hour). Speed was altered as required and the boat was stopped frequently to
search complex areas with binoculars and to obtain counts of the number of animals in rafts.
During small boat surveys, rafts of sea otters were counted using 7X50 binoculars and 14X40
stabilized binoculars. Female rafts were distinguished from male rafts by the presence of pups.
Rafts were counted by all observers. Each person assessed the raft size independently, making
several counts when possible and then compared counts. The final count accepted was the best
overall repeatable count and this value was often among the higher of the initial counts,
because counts were made as the boat and otters rose and fell in the ground swell and high
counts were obtained when an observer achieved an unobstructed view of the sea otter raft.
Additional counts were collected in 2009 and 2013 by observation from a large ship following a
report of a raft of sea otters (G. Ellis pers. com. 2008) on Cook Bank, offshore of the Cape Scott
to Hope Island segment of northern Vancouver Island. To survey the area, two observers
collected sightings from the top viewing deck in 2013, (10 m above the water), and from the
bridge in 2009 (7.5 m above the water) of the Canadian Coast Guard vessel Tanu as it made a
single transit travelling parallel to the coast at 10 knots. The observers scanned the ocean on
either side of the vessel and the position of otter sightings were approximated as the position of
the ship at the time of the sighting.
Sea and weather conditions were recorded during surveys and were categorized as follows.
Good to Excellent survey conditions existed when sea state ranged from flat calm (Beaufort 0)
to swells up to 1 m and wind speeds less than 10 knots (18 km/hr) (Beaufort 3). Fair conditions
were defined generally as seas 1 to 1.5 m or when wind speeds ranged from 10 to 15 knots (28
km/hr) (Beaufort 4). Poor conditions were generally defined as seas greater than 1.5 m or wind
speeds greater than 15 knots (28 km/hr). High overcast created ideal lighting conditions due to
reduced sun glare. Surveys did not commence in Poor conditions or when visibility was
obscured particularly through binoculars, e.g., by rain or fog. Although Good to Excellent
conditions were sought, sea states often changed during the course of a survey. If conditions
deteriorated, the survey was repeated at a later date whenever possible. All surveys to
document minimum population size, distribution and range of occupation have been carried out
from May through early September since 2004. In earlier years surveys were conducted
between April and September. Further details about the survey methodology including details
about the methodology adapted for use from helicopter (some surveys 2001 to 2004) are
described in Nichol et al. (2005).
ANALYTICAL METHODS
Annual survey segment counts made in 2013 and counts from new areas occupied by rafts of
sea otters were summed to obtain a minimum population size for 2013 for the Vancouver Island
region and the central mainland coast region. To estimate growth rates in the population over
time, we first fit growth models to annual survey data on a sub-regional basis. Survey segments
were grouped into sub-regions according to similarity in years of occupation and contiguous
4
geography. Annual survey counts from the member segments were summed. Grouping survey
segments into sub-regions also achieved a minimum of three years of survey effort over a fiveyear period, which was needed to fit models. The sub-regions do not represent distinct
population units, but this approach made it possible to assess growth trends on a smaller scale
and allowed incorporation of much more survey data from years between the range wide
surveys during which only portions of the sea otter range were surveyed. The 16 segments that
comprise the Vancouver Island region were grouped into seven sub-regions. The six segments
in the central mainland coast region were grouped into three sub-regions. To obtained a
regional estimate of growth rate, that could be compared with the regional growth rate estimates
from previous assessments, we combined the sub-regional model estimates.
The time series for each segment in a sub-region was inspected for gaps and for multiple
surveys. If more than one survey had been made in a year in a segment, the survey made
under the best conditions, which generally yielded the higher count, was selected. If a survey
had not been completed in a segment in a year, then that year was excluded from sub-region
time series, unless the missing count would make up < 20% of the total sub-region count. In
these instances we estimated the number of animals in a missed segment by interpolation using
the exponential equation that best fit a maximum of four counts preceding and/or succeeding
the missing count. Only four interpolated values were used among all sub-regions and years.
In this assessment, we fit two alternative growth models to the time series available for each
sub-region using Maximum Likelihood methods. All calculations were performed using MatLab
R2014a. The models evaluated were for exponential and for logistic population growth with
discrete 1-year time steps, described by the following equations:
(1)
(2)
exponential: =
N pred ,t N pred ,t −1 ⋅ e r
logistic: =
N pred ,t N pred ,t −1 ⋅ e
 N

r 1−  pred ,t −1  
K

 
where Npred,t is the predicted abundance at time t, and r (the intrinsic or instantaneous rate of
growth) and K (carrying capacity) are fitted parameters. The solution to both models required
one additional estimated parameter, N0, the initial abundance at year 0 (defined as the year
before the first available count for that subregion). Model fitting was accomplished using
unconstrained numeric optimization to maximize the following likelihood function (Hilborn and
Mangel 1997) which compared observed to predicted counts:
(3)
N 0,i )
( N obs | Model j , ri , ki , =
1
2⋅p ⋅s 2
⋅e
 − ( N obs − N pred )2 


2


2
⋅
s


where Nobs is the vector of observed counts and Npred is the vector of predicted counts, given a
particular model and set of parameter estimates (ri, Ki and N0,i). The parameter s2 represents
the variance term, solved for by iterative re-weighting (following Pascual et al. 1997). We thus
assume that deviations between observed and model-predicted counts were normally
distributed with mean 0 and standard deviation s, and thus we assume that deviations primarily
reflect observer error rather than process error (Hilborn and Mangel 1997; Tinker et al 2006).
We used the Akaike Information Criterion (AIC) to select the most parsimonious of the two
models for each sub-region, and calculated R2 to evaluate goodness of fit of the selected model.
For each sub-region we report maximum likelihood estimates and associated standard errors for
5
r and K (the latter is reported only for those sub-regions where logistic growth was the best
supported model), and also for s (expressed as a coefficient of variation, CV = s/N).
Additionally, for each sub-region we report the estimated finite rate of growth (rd) for the most
recent 5-year period (2009-2013) and for the previous 5-year period (2004-2008). Note that rd,t
is a measure of the average realized rate of population growth over some defined interval
(expressed in units of annual percent increase in population size) and thus differs from the
model parameter r (the instantaneous rate of growth as N →0). We calculated rd,t as:
(4)
1


4
N


pred ,t

=
− 1
rd ,t 100 

N


pred ,t − 4 


We calculated a standard error for rd,t using a resampling approach, whereby we generated
20,000 random parameter sets for equation 1 or 2 (as appropriate), used these to solve
equation 4, and calculated the standard deviation of rd,t across all iterations. Parameter sets for
each iteration were drawn from a multivariate random normal distribution with mean values
equal to the maximum likelihood estimates and variance-covariance matrix calculated as the
inverse of the hessian matrix derived from the maximum likelihood solution of equation 1 or 2.
We also report rd,t for the entire region (i.e., by summing the annual expected counts derived
from the best fit model for each sub-regions).
Maps
Map figures were produced using ArcMap 10.2.2. Survey effort is represented by kilometres of
survey transect and sea otter distribution by the location of sightings in 2008 and in 2013.
Numbers of sea otters per sighting are not depicted on the map figures.
RESULTS
REGIONAL POPULATION ESTIMATES AND GROWTH RATES
Surveys of the Vancouver Island region were made from May 13 to August 25, 2013. The
central mainland coast region was surveyed from May 9 to May 21, 2013 and from June 10 to
June 22, 2011. Survey conditions during both 2013 and 2008 (for comparison) are presented in
Table 1. Range expansion has taken place since 2008 in both regions (Figure 3). The
geographic distribution of survey effort and otter sightings in 2013 is presented in Figures 4, 5,
and 6 and in 2008 in Figures 7, 8 and 9. Sea otters were observed singly and in groupings of up
to 250 animals. Approximately half of the sea otters were observed in rafts of more than 30
animals, the other half in smaller groups and as single animals (Figure 10).
Vancouver Island Region
A total of 5,612 sea otters were counted in the Vancouver Island region in 2013 (Table 2). The
finite rate of growth for the region was 7.12% year-1 (SE = 1.05) over the period of 2009 to 2013
inclusive. During the five-year period 2004 to 2008, the finite rate of growth was 6.77% (SE =
1.20). These estimates are not significantly different from each other (Figure 11).
Central Mainland Coast Region
A total of 1,142 sea otters were counted in 2013. In 2011, 757 sea otters were counted
although the Aristazabal Island segment and the Seaforth Channel to Ivory Island and Lady
Douglas Island segment were not surveyed in that year. The finite rate of growth was estimate
6
to be 12.56% year-1 (SE = 1.65) from 2009 to 2013 and 4.22% year-1 (SE = 1.62) from 2004 to
2008 (Figure 12).
SUB-REGIONAL ANNUAL RATES OF INCREASE AND RANGE EXPANSION
Vancouver Island Region
Rates of growth among sub-regions generally corresponded inversely with years of occupation
and with the proportion of the total population in sub-regions. Near the centre of the range,
where the population has been established for the longest period of time (sub-regions 5 and 6)
the annual rates of increase were the lowest of all the sub-regions on Vancouver Islands in both
the 2004 to 2008 and the 2009 to 2013 periods. In 2009 to 2013 the growth rate in sub-region 5
was 2.57% year-1 (SE = 0.66), and in sub-region 6, 2.54% year-1 (SE = 1.23). A logistic
population growth model, best describes the pattern of growth in these sub-regions where half
of all sea otters on Vancouver Island occur. Rates of annual increase are higher among subregions to the south and to the north and three northern sub-region most recently occupied had
average annual rates of increase during 2009 to 2013 that exceeded 20% year -1 (Figure 13 and
Table 3).
Since 2008 range expansion events in the Vancouver Island region have taken place mainly in
the north in Queen Charlotte Strait (see Figures 3, 4, 7). The presence of a raft of male sea
otters amongst an island group in Queen Charlotte Strait east of Hope Island was recorded on
surveys in 2009 and 2010. A raft of males was observed along the mainland shore of Queen
Charlotte Strait in 2013 and another raft was observed north of Queen Charlotte Strait in Smith
Sound in 2013. In contrast there has been little evidence of range expansion elsewhere,
although sea otters expanded into Esperanza Inlet in 2008 and since then rafts of male sea
otters have been observed in adjoining Tahsis Inlet (R. Dunlop, biologist for Nuu-chah-nulth
Tribal Council, Gold River, BC pers. comm. 2013) indicating that sea otters are now established
around all of Nootka Island. An example of range expansion and subsequent occupation and
population growth is illustrated in Figure 14 with survey data from the Catala Island segment.
Initial immigration of a raft of male sea otters in 1992 was followed, in subsequent years, by
population growth as females established in the area.
Central Mainland Coast
Sub-region 10 which includes the Goose Group segment and the McMullin Group to Cape Mark
segment has been occupied for longer than any other part of the central coast range. More than
half of the otters counted on the central mainland coast in 2013 occurred in this sub-region.
Growth in this sub-region was minimal and similar to the longest occupied sub-regions of
Vancouver Island, with an annual rate of increase in 2009 to 2013 of 2.41% year-1 (SE = 1.33).
In contrast the sub-region to the north that includes Aristazabal Island (sub-region 8) and subregion 9 to the east, both of which have only been occupied since no earlier than 2007 have
high annual rates in excess of Rmax (Figure 15 and Table 4).
Range expansion to Calvert Island occurred after the surveys were completed in June 2013 (L
Honka, M.Sc. student, School of Resource and Environmental Management, Simon Fraser
University pers. comm. 2013). Although not observed during surveys, a large raft of male sea
otters arrived during summer and therefore the west side of Calvert Island was considered
occupied in 2013. Elsewhere in sub-regions 9 rafts of females were recorded by 2011.
7
DISCUSSION
The sea otter population in British Columbia has continued to grow and its distribution increase
demonstrating population growth patterns that are typical of recovering sea otter populations.
Annual growth rates were much lower in long occupied sub-regions of Vancouver Island and the
central mainland coast than in more recently occupied sub-regions reflecting density
dependence, whereby growth approaches zero as population size approaches carrying capacity
(K). Growth rates in more recently occupied sub-regions were higher and in many cases higher
than Rmax estimates for the species, indicating that immigration was an additional factor (Estes
1990).
The survey methodology provides an index of abundance assumed to represent a constant
proportion of the population. Sea otters are a non-migratory nearshore species. They exhibit raft
site fidelity and occupy small coastal home ranges over their lives. These attributes contribute to
the success of the direct count method. Similar methods have been used to assess trends in
population size and growth in Washington State and California (Jameson and Jeffries 2014;
USGS 2014). Data collected with this survey method in BC also provide detailed, relatively finescale information about the distribution of sea otters, locations of rafts by sex and the timing of
range expansion events, because parts of the range are surveyed annually in addition to the
range wide survey completed at five-year intervals.
A challenge for long-term population survey programs such as this, however, is maintaining a
consistent level of effort as the population increases and expands its range. In response to this
challenge the survey method was adapted in the early 2000’s for surveys from helicopter, a
platform that would allow more rapid coverage of survey segments and thereby more survey
coverage per unit time. Counts obtained from boat and from helicopter made in the same survey
segment and within the same season produced comparable results. This supported the
assertion that counts by the two methods could readily be used together in analysis (Nichol et
al. 2005). However, helicopter surveys are significantly more expensive and have not been
possible in recent years but the challenge of maintaining consistent effort persists.
In 2009 and during the range wide survey in 2013, sea otters were counted from the CCG ship
Tanu on an offshore transect as well as by small boat along the established inshore route at the
north end of Vancouver Island in the Cape Scott to Hope Island segment. The offshore area had
not previously been surveyed. The resulting counts are used in the growth rate estimates and in
the population index count for 2013 reported here. It is unknown, however, how many sea otters
might have been counted in the offshore transect had it been surveyed in earlier years. It will be
important to continue to incorporate offshore survey effort over this shallow bank in the future.
Small boats are well suited for surveys in shallow rocky nearshore habitat, where a shallow draft
and maneuverability are needed. Large ships are better suited for survey of expansive offshore
areas because of greater efficiency and stability, and the higher viewing platform provides a
wider field of view. Further methodological development is needed though to be able to
compare survey results from ships and small boats as was done previous with counts from
small boats and helicopters (Nichol et al. 2005).
Regional annual growth rate estimates for Vancouver Island and the central mainland coast, as
a whole, were generally consistent with previous growth trajectories for the two regions (Nichol
et al. 2005; 2009). The Vancouver Island regional growth trends for the two recent five-year
intervals, 2004 to 2008 and 2009 to 2013 were similar but slightly lower than the 8.4% year -1
estimate previously for 1995 to 2008 (Nichol et al. 2009). This probably reflects density
dependence in growth in the long occupied sub-regions, where logistic population growth
models fit best These sub-regional trends were not as well represented in previous
assessments that estimated growth from one exponential growth model fitted to the entire
region, over a longer time series but with a more limited data set.
8
The central mainland coast regional growth rate has always been below Rmax for the species
(Nichol et al. 2005, 2009). Although it is unclear why this has been the case, it is possible that
greater variability in early counts in the time series, when there was less knowledge of the
distribution of sea otters in this area, masked an early phase of rapid growth. Previous
assessments have noted a poorer fit of the linear regression to central coast counts compared
to Vancouver Island counts, which would suggest greater variability among counts. In this
assessment, the width of the confidence intervals around the central mainland coast regional
growth curve estimates for the entire time series from 1990 to 2013 are wider than those for the
Vancouver Island growth curve estimate and therefore continue to reflect this uncertainty.
We detected significantly different annual growth rates on the central mainland coast between
the two five-year time periods, 2004 to 2008 and 2009 to 2013. The much higher growth rate
estimate in the latter period includes the contribution from two additional sub-regions of the
central mainland coast only recently occupied (since 2007). In the previous assessment the
average annual growth rate for the entire region based on a single exponential population
growth model fit to data from 1990 to 2008 did not capture the sub-regional effects (Nichol et al.
2009).
Fitting growth models to sub-regional data sets allowed for use of more years of survey data
than in previously assessments. In previous BC sea otter population assessments (e.g., Watson
et al. 1997; Nichol et al. 2005; Nichol et al. 2009), a single regional growth rate for each region
was estimated from an exponential growth curve, fitting log counts on year (time) in a linear
regression analysis fit by least squares (Zar 1984). As such the previous trend analysis was
restricted to use of survey data from years in which an entire region (Vancouver Island or
central mainland coast) had been surveyed (or ≥ 70% of the region, Nichol et al. 2005, 2009)
(see Table 2). This meant excluding many years of survey effort in which only some segments
had been surveyed. Fitting growth models to sub-regional data sets has allowed us to use
survey data collected during the five years between 1995 and 2001, and the seven years
between 2004 and 2013. The result has been a more detailed assessment of the pattern of sea
otter population growth in BC.
Although sub-regions do not represent distinct population units, analysis at this level serves to
better illustrate the process of sea otter population growth that is a result of demographic
processes and range expansion events. Sea otter populations are limited by the availability of
food. Growth in the population is a result of births and deaths, measurable on an annual basis,
but also as a result of immigration and emigration, which manifest as range expansion events
that are detected at longer and irregular intervals. Although food limitations result in density
dependent population growth and range expansion, the timing of range expansion events is
influenced by social dynamics, the distance from occupied areas to suitable new habitat and the
quality of habitat (Tinker et al. 2008; Lafferty and Tinker 2014). Gregr et al. (2008) modelled
habitat complexity (a proxy for habitat quality for sea otters) along the west coast of Vancouver
Island from the southern tip of Vancouver Island to a point to the west of Hope Island, and
identified habitat areas of similar complexity to those in Checleset Bay, Mission Group and
Kyuquot Sound, the areas that comprise the longest occupied sub-region (sub-region 5) in our
assessment. Within this study area the results suggested that there may be less similarly
complex habitat south of Estevan Point to Clayoquot Sound, (although lots of habitat in
Clayoquot Sound) and less complex habitat from Quatsino Sound north. Thus if habitat quality
and distribution influences range expansion, then this may partly explain the distribution of sea
otters amongst the sub-regions. In other words, the quantity and quality of “good” habitat may
be lower outside the two sub-regions where most of the otters are found and this slowed the
process of range expansion. Presumably this could change in the future as the population
expands further into different habitat which might involve “good” areas.
Annual growth rates during the period 2009 to 2013 were generally at or above Rmax in recently
occupied sub-regions in both the Vancouver Island region and the central mainland coast
9
region, with the exception of the most southerly sub-region on Vancouver Island (sub-region 7).
In sub-region 7 of Vancouver Island that includes Clayoquot Sound, the rate of annual increase
from 2004 to 2013, (8.98% year-1, SE = 1.82) although higher than the regional rate for
Vancouver Island, seems low in an area that should represent the expanding southern edge of
the range and an area that according to habitat modelling should have a lot of suitable habitat
(Gregr et al. 2008). Range expansion by males into Clayoquot Sound was recorded in 2004 but
rafts of females were first observed in 2010, suggesting as much as a six-year interval between
range expansion by males and the arrival by females.
In contrast, annual rates of increase in the three most northern sub-regions of Vancouver Island
were above Rmax. In sub-region 2 (Cape Scott to Hope Island), for example, the rate of increase
estimated from 2004 to 2013 is attributable to increasing numbers of female and pups counted
on surveys in the area near Hope Island over the time series. Sea otters observed on the
offshore transect in this sub-region in 2013 were spread out foraging, suggesting the offshore
area is a productive foraging habitat in close proximity to nearshore rafting areas.
On the central mainland coast, range expansion to Aristazabal Island was first documented on a
survey in 2008. This expansion left a geographic gap in the range between the sea otter
occupied range on the south side of Milbanke Sound and the occupied range at Aristazabal
Island. The gap between these two areas persisted in 2013. Range expansion with gaps
between occupied areas in California has been attributed to variation in habitat quality and this
is a plausible explanation here as well (Lubina and Levin 1988; Reidman and Estes 1990;
Tinker et al. 2008). Historical reports of sea otters in inland passages 2 of BC suggests gaps will
be filled as prime habitat reaches equilibrium densities. Indeed this may be occurring on
Vancouver Island where recent occupation of inlet habitat such as Esperanza Inlet and
Quatsino Sound by male rafts since 2001 may be an example of occupation of lesser quality
habitat as a trade-off to maintain proximity to females (Nichol et al. 2009). On a larger scale, the
gap that has persisted between the Vancouver Island region and the central mainland region is
now disappearing as sea otters on the central mainland coast have expanded southward to
Calvert Island and are also now present in Smith Sound and in Queen Charlotte Strait. In future
assessments Vancouver Island and the central mainland coast may need to be viewed as one
large region.
The BC sea otter population continues to grow and recover from extirpation in the early part of
the 20th century. There are no estimates of the size of the BC sea otter population prior to the
maritime fur trade, against which to measure recovery. There are various incomplete historical
records that suggest a much larger population than today (Nichol 2015). For example, on Haida
Gwaii a single ship trading for less than two months in 1787 amassed 1,821 pelts and another
obtained 1,400 sea otters in less than two months in 1791 (Sloan and Dick 2012). From 1790 to
1800 an estimated 10,000 pelts a year were obtained by American vessels alone trading with
native people along the coast of northern British Columbia and southeastern Alaska
(Dmytryshyn and Crownhart-Vaughan 1976). These statistics suggest that high densities of sea
otters were once found not only along Vancouver Island, but also around the islands of Haida
Gwaii and along the central and northern mainland BC coast.
Gregr et al. (2008) provided several possible estimates of a maximum sea otter population for
BC based on a habitat modeling approach. One model for the west coast of Vancouver Island
estimated a maximum population of 5,123, (95% CI=3,337–7,104), in high quality habitat and
14,831, (95% CI=9,790–20,751) coast-wide. A second model suggested, 52,459 sea otters
coast-wide (95% CI = 34,264–73,489). On the west coast of Vancouver Island, the population
2
Menzies, naturalist aboard the HMC Discovery in 1790 reports a sighting of a large raft of sea otters in
the inland waters of Wright Sound which is north east of Aristazabal Island
10
count is now within the range of model estimate, but coast-wide the population is far below the
estimates of both models.
ACKOWLEDGEMENTS
BC sea otter population surveys in 2013 would not have been possible without the contributions
of many colleagues. We thank the following observers in 2011, L. Honka, and in 2013, C. Blain,
M. DeRoos, L. Honka, S. Hutchings, L. Saville and R. Dunlop and the Nuu-chah-nulth Tribal
council. We thank the captain and crew of the CCGS Tanu for their assistance with northern
Vancouver Island surveys, the Port Hardy Lifeboat Station, Bella Bella Fisheries Officers, and
M. Barbeau, R. Snyder, and K. Hansen for other logistical support. John Ford provided valuable
review comments and edits to both the original working paper in October 2014 and to the final
research document. Tim Tinker provided valuable input as the external reviewer of the original
working paper presented at the National Marine Mammal Peer Review meeting in Ottawa in
October 2014. For this final version he provided analytical assistance with revisions to the
population growth rate analysis. Funding was provided by the Species-at-risk Program,
Department of Fisheries and Oceans.
REFERENCES
Bigg, M.A. and I.B. MacAskie. 1978. Sea otters re-established in British Columbia. J. Mammal.
59: 874-876.
Bodkin, J.L., G.G. Esslinger and D. Monson. 2004. Foraging depths of sea otters and
implications to coastal marine communities. Mar. Mamm. Sci. 20 (2): 305-321.
Dmytryshyn, B., Crownhart-Vaughan, E.A.P., 1976. Kyrill T. Khlebnikov’s reports, 1817-1832.
Colonial Russian America. Translated with Introductions and Notes by B. Dmytryshyn and
E.A.P. Crownhart-Vaughan. Oregon Historical Society, Portland.
Estes, J. A. 1990. Growth and equilibrium in sea otter populations. J. Anim. Ecol. 59: 385–401.
Ford, J.K.B. 2014. The Marine Mammals of British Columbia. Royal British Columbia Museum
Handbook. Volume 6.
Garshelis, D.L., and Garshelis, J.A., 1984, Movements and management of sea otters in
Alaska: J. . Wildl. Manag., v. 48, p. 665-677.
Garshelis, D.L., A.M. Johnson, and J.A. Garshelis. 1984. Social organization of sea otters in
Prince William Sound, Alaska. Can. J. Zool. 62:637-647.
Gerber, L. R., T. Tinker, D. Doak, and J. Estes. 2004. Mortality sensitivity in life-stage simulation
analysis: a case study of southern sea otters. Eco. Appl. 14: 1554–1565.
Gregr, E. J., L. M. Nichol, J. C. Watson, J. K. B. Ford, and G. M. Ellis. 2008. Estimating carrying
capacity for sea otters in British Columbia. J. . Wildl. Manag. 72:382–388.
Hilborn, R., and M. Mangel. 1997. The ecological detective: confronting models with data.
Princeton University Press, Princeton, NJ.
Jameson, R.J., 1989, Movements, home ranges, and territories of male sea otters off central
California. Mar. Mamm. Sci., 5: 159-172.
Jameson, R.J., K.W. Kenyon, A.M. Johnson, and H.M. Wright. 1982. History and status of
translocated sea otter populations in North America. Wild. Soc. Bull. 10:100-107.
Jameson, R.J. and S. Jeffries. 2014. Results of the 2013 survey of the reintroduced sea otter
population in Wahsington State. Washington Department of Fish and Wildlife. Wildlife
Science Program. Marine Mammal Investigations. 7p
11
Lafferty, K. D., and M. T. Tinker. 2014. Sea otters are recolonizing southern California in fits and
starts. Ecosphere 5(5):art50.
Lubina J.A. and S.A. Levin 1988. The spread of a reinvading species: range expansion in the
California sea otter. The Amer. Nat. 131(4): 526-543.
MacAskie, I.B. 1987. Updated status of the sea otter (Enhydra lutris) in Canada. Can. Field-Nat.
101:279-283.
Morris, R.D., D.V. Ellis, and B.P. Emerson. 1981. The British Columbia transplant of sea otters
Enhydra lutris. Biol. Cons. 20:291-295.
Nichol, L.M. 2015. Conservation in Practice. In: Sea Otter Conservation Eds: S. Larson, G.
VanBlaricom and J. Bodkin. Elsevier.
Nichol, L.M.,Boogaards, M.D., Abernethy, R. 2009. Recent trends in the abundance and
distribution of sea otters (Enhydra lutris) in British Columbia. DFO Can. Sci. Advis. Sec.
Res. Doc. 2009/016. iv + 16 p.
Nichol, L. M., J. C. Watson, G. E. Ellis, and J. K. B. Ford. 2005. An assessment of abundance
and growth of the sea otter (Enhydra lutris) population in British Columbia. DFO Can. Sci.
Advis. Sec. Res. Doc. 2005/094. ii + 22pp.
Pascual, M. A., P. Kareiva, and R. Hilborn. 1997. The influence of model structure on
conclusions about the viability and harvesting of Serengeti wildebeest. Conservation
Biology 11:966-976.
Ralls, K., Eagle T., and Siniff, D.B., 1988, Movement patterns and spatial use of California sea
otters, in Siniff, D.B., and Ralls, K., eds., Final Report on Contract No. 14-12-001-3003,
Population status of California sea otters: Minerals Management Service, Los Angeles,
CA, p. 33-63.
Riedman, M.L., and J.A. Estes. 1990. The sea otter (Enhydra lutris): Behaviour, ecology and
natural history. US Fish and Wildl. Serv. Biol. Rep. 90(14). 126 pp.
Sloan, N.A., Dick, L., 2012. Sea Otters of Haida Gwaii: Icons in Human-Ocean Relations.
Archipelago Management Board and Haida Gwaii Museum, 184 pp.
Thometz, N. M., M. T. Tinker, M. M. Staedler, K. A. Mayer and T. M. Williams. 2014. Energetic
demands of immature sea otters from birth to weaning: implications for maternal costs,
reproductive behavior and population-level trends. J. Exp. Biol. 217: 2053-2061
Tinker, M. T., D. F. Doak, and J. A. Estes. 2008. Using demography and movement behavior to
predict range expansion of the southern sea otter. Eco. Appl. 18:1781–1794.
Tinker, M. T., D. F. Doak, J. A. Estes, B. B. Hatfield, M. M. Staedler, and J. L. Bodkin. 2006.
Incorporating diverse data and realistic complexity into demographic estimation
procedures for sea otters. Eco. Appl. 16:2293–2312.
Watson, J.C., G.M. Ellis, T.G. Smith, and J.K.B. Ford. 1997. Updated status of the sea otter,
Enhydra lutris, Canada. Can. Field-Nat. 111(2): 277-286.
United States Geological Survey (USGS) 2014. Spring 2014 California Sea Otter Survey.
United States Fish and Wildlife Service (USFW) 2014. Stock Assessment Report. Sea otters
(Enhydra lutris): Southeast Alaska stock 18pp.
Zar, J. H. 1984. Biostatistical Analysis. 2nd Edition. Prentice-Hall Inc. New Jersey.
12
TABLES
Table 1. Kilometres of survey effort and survey conditions in 2008 and 2013.
Region
Year
Kilometres
Kilometres surveyed by conditions in all
areas surveyed (%)
Excellent
Good
Fair
Poor
Vancouver
Island
2013
1931
767 (40)
605 (31)
395 (20)
165 (9)
Vancouver
Island
2008
1298
196 (15)
732 (56)
287 (22)
83 (6)
Central
mainland
2013
963
0
350 (36)
523 (54)
89 (9)
Central
mainland
2008
839
157 (19)
313 (37)
320 (38)
46 (6)
13
Table 2. Sea otter counts by region in years when at least one of the regions was surveyed entirely*.
Platforms : fixed-wing aircraft- 1, helicopter- 2, small boat- 3. Coast Guard ship - 4. BC Total is only
given when both regions were surveyed.
Year
Vancouver
Island
Central Mainland
Coast
BC Total
Platforms
1977
1978
1980
1982
1984
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1998
2001
2002
2003
2004
2007
2008
2013
70
67
74
116
345
370
354
582
612
486*
801*
910
1188
1423
2673
2134*
2415
2765*
4110
5612**
56
104
168
135
259
429
507
235
394*
420
642
602
1142
70
67
74
116
345
370
354
582
668
590
969
1045
3180
2369
2809
3185
4712
6754
1
1
1
1
1,3
1
3
2,3
2,3
3
2,3
3
3
3
3
3
2,3
2,3
2,3
2,3
3
3
3,4
* includes a few interpolated values for missed counts that total a small proportion of the regional count
(see Table 1 in Nichol et al. 2009).
** 506 animals counted in sub-region 2 on inshore transect, 396 counted on offshore transect from CCG
ship Tanu.
- region not yet occupied
Source of survey data displayed in Table 2
1977: Bigg and MacAskie 1978
1984: MacAskie 1987
1978: Morris et al. 1981
1987-1995: Watson et al. 1997
1980-82: Watson et al. 1997
1996-2013: DFO, J. Watson
14
Table 3. Model fitting results for sub-regions of Vancouver Island illustrated in Figure 13.
Sub-region
Colour
on
Figure
14
Date
range
Survey
years N
AIC selected
model
2009-2013
annual
growth rate
(SE)
2004-2008
annual
growth rate
(SE)
Instantaneou
s rate of
increase
(SE)
Estimate of
Carrying Capacity
K (SE)
CV
R
2
1. (Queen Charlotte
Strait east and
Smith Sound)
2009 2013
3
Exponential
25.77 (9.46)
--
0.29 (0.12)
--
0.63
0.77
2. (Cape Scott to Hope
Island)
2001 2013
8
Exponential
29.12 (1.46)
29.12 (1.46)
0.26 (0.01)
--
0.13
0.99
3. (Scott Islands)
2008 2013
3
Exponential
23.69 (1.57)
--
0.21 (0.01)
--
0.04
0.99
4. (Brooks Bay,
Quatsino Sound,
Kains to Cape
Scott)
1989 2013
14
Logistic
(AIC, L = 182.03,
E = 185.09)
4.84 (2.87)
9.37 (3.98)
0.20 (0.06)
1197.81 (247.68)
0.39
0.87
5. (Checleset Bay,
Mission Group,
Kyuquot Sound)
1977 2013
32
Logistic
(AIC, L = 389.14,
E = 409.97)
2.57 (0.66)
4.54 (0.65)
0.14 (0.02)
2083.56 (194.51)
0.22
0.96
6. (Nootka Island
Nuchatlitz Inlet,
Catala Island,
Esperanza Inlet)
1977 2013
22
Logistic
(AIC, L = 263.65,
E = 275.25)
2.54 (1.23)
4.54 (1.75)
0.15 (0.02)
1014.82 (125.49)
0.32
0.92
7. (Hesquiat
Peninsula,
Clayoquot Sound)
1995 2013
9
Exponential
8.98 (1.82)
8.98 (1.82)
0.08 (0.02)
--
0.29
0.77
15
Table 4. Model fitting results for sub-regions of the central mainland coast, illustrated in Figure 15.
Colour
on
Figure
15
2009-2013
annual
growth rate
(SE)
2004-2008
Annual
growth rate
(SE)
Instantaneou
s rate of
increase
(SE)
Estimate of
Carrying Capacity
K (SE)
Date
range
Survey
years N
AIC selected model
8. (Aristazabal Island
and Seaforth to
Ivory and Lady
Douglas Island)
2007 2013
3
Exponential
31.97 (0.61)
--
29.85 (0.11)
0.02
0.99
9. (Simmonds Group
to Kildidt Sound)
2007 2013
4
Exponential
35.06 (3.62)
--
27.57 (2.08)
0.18
0.97
10. (Goose Group,
McMullin Group to
Cape Mark )
1990 2013
13
Logistic (AIC,
L = 156.63,
E = 157.90)
2.41 (1.33)
4.22 (1.62)
2.01 (1.23)
0.24
0.84
Sub-region
16
709.11 (174.61)
CV
R
2
FIGURES
Figure 1. Survey region boundaries and areas within that were surveyed in 2008 and 2013. Checleset
Bay site of the original re-introductions 1969-1972. Goose Group where groups of sea otters were first
sighted on the central mainland coast in 1989.
17
Figure 2. Survey segments and new areas surveyed in 2013. Survey segments: 1. Clayoquot Sound, 2.
Hesquiat Peninsula, 3. Nootka Island, 4. Nuchatlitz Inlet, 5. Catala Island, 7. Kyuquot Sound, 6.
Esperanza Inlet, 8. Mission Group, 9. Checleset Bay, 10. Brooks Bay, 11. Quatsino Sound, 12. Kains
Island to Cape Scott, 13. Scott Islands , 14. Cape Scott to Hope Island, 15. Queen Charlotte Strait east,
16. Calvert Island, 17. Kildidt Sound, 18. Simond Group to Tribal Group, 19. Goose Group, 20. McMullin
Group to Cape Mark, 21. Seaforth to Ivory Island and Lady Douglas Island, 22. Aristazabal Island. New
areas surveyed to assess range expansion: a. Inlets connected to Esperanza Inlet, b. Holberg Inlet, c.
Smith Sound, d. Price Island, e. extension of Aristazabal Island area.
18
Figure 3. Sea otter range in British Columbia in 2013 and place names mentioned in the text.
19
Figure 4. Distribution of sea otter sightings and survey effort on northern Vancouver Island in 2013.
20
Figure 5. Distribution of sea otter sightings and survey effort on central west coast Vancouver Island in
2013.
21
Figure 6. Distribution of sea otter sightings and survey effort on the central mainland coast in 2013.
22
Figure 7. Distribution of sea otter sightings and survey effort on northern Vancouver Island in 2008.
23
Figure 8. Distribution of sea otter sightings and survey effort on the central west coast of Vancouver
Island in 2008.
24
Figure 9. Distribution of sea otter sightings and survey effort on the central mainland coast in 2008.
25
Figure 10. Number of sea otters by group sizes that comprised the 6,754 sea otters counted during
surveys in 2013.
26
A.
B.
Figure 11. A. Estimated population growth curve in the Vancouver Island region 1977 to 2013 on a log
scale. Solid line represents the modelled expected values. Dotted line represents approximate 95%
confidence intervals. Diamonds represent survey counts from years when all occupied survey segments
were surveyed. B. Growth curve and 95% confidence intervals and survey count on an ordinal scale.
27
A.
B.
Figure 12. A. Estimated population growth curve on the central mainland coast region 1990 to 2013 on a
log scale. Solid line represents the modelled expected values. Dotted line represents approximate 95%
confidence intervals. Diamonds represent survey counts from years when all occupied survey segments
were surveyed. B. Growth curve and 95% confidence intervals and survey count on an ordinal scale.
28
Figure 13. Estimated growth trends in the Vancouver Island region. Inset graphs show growth curves by
subregion, either logistic or exponential (on a log scale) fitted to sea otter survey data. Inset pie chart
gives the proportions of the 2013 count ( 5,612 sea otters) by sub-region.
29
Number of sea otters
300
250
200
150
100
50
0
1975
1985
1995
Year
2005
2015
Figure 14. Pattern of colonization and growth observed in the Catala Island segment following initial
range expansion by a raft of males in 1992 (red triangle). Diamonds represent counts in subsequent
years which were predominantly females and pups.
30
Figure 15. Estimated growth trends in the central mainland coast region. Inset graphs show growth
curves by subregions either logistic or exponential (on log scale) fitted to sea otter survey data. Inset pie
chart gives the proportions of the 2013 count (1,142 sea otters) by sub-region.
31